Supplier Quality - What constitutes a Reject?

R

RosieA

My company has 12 plants and each plant has it's own methods for evaluating supplier quality. I'm attempting to address that now, and get us using common metrics, but I am finding dramatic differences in how we determine what a reject is. We're all in the same market: industrial (vs. a regulated marketplace)

Plants A, B, and C are not doing routine receiving inspection, but have determined supplier quality based on in-process problems, attributable to the supplier. The ppm rate for these plants runs from 4500-9000.

Plants D, E and F are doing receiving inspection and their ppm rates range from 30,000-60,000ppm.

These are some pretty dramatic differences.

Plants who do receiving inspection reject for a number of non-product related problems, such as a missing cert or paperwork error.

I'd be interested in knowing how other companies have defined what a supplier reject is, and whether it includes administrative issues that do not affect the useability of the product supplied.

Note: I am not advocating one method over another, simply trying to understand how best to reflect a common picture from plant to plant.

Thanks
 
Q

quality1

Re: Supplier Quality--what constitutes a reject?

We deal strictly with product inspection (form, fit, function), ocassionally paperwork would be missing or wrong. The supplier would send the missing or correction electronically to us when we ask for it. I think if you have a trend, you need to send the supplier your expectation on required documentation and a CPAR to address it.
 
S

Sorin

Re: Supplier Quality--what constitutes a reject?

Usually I set administrative errors separate.
Just to be clear, by administrative errors I mean missing papers not wrong papers (if I am asking integral test bars and the CoC state separate test bars, that's a reject on the spot),

ON the other hand, I heard from a prime that 30 - 40% of their rejections (REAL rejections) are clerical errors.
 
T

tomvehoski

Re: Supplier Quality--what constitutes a reject?

There is no easy answer or strict definition, usually. It is also difficult to impossible to boil it all down to one number.

I have parts sitting on my desk right now from a supplier that are not to the print. The print does not have any key characteristics on it, but our engineers know what is important. There are also some typos on the print in dimensions. We kicked off the supplier late and pushed them for parts even though we didn't give them time to finalize tooling and send in samples for approval. We never discussed ahead of time how we would measure, what we would measure, etc. Is that a supplier reject, project management reject, engineering reject or ????? In reality it has several root causes.

I prefer to handle things as exceptions rather than trying to come up with complicated scoring methods. If I keep having recurring issues with someone, then it gets handled in a workshop format vs. nasty form letters and threats of lost business. Not everything that you can measure SHOULD be measured...
 
S

Sorin

Re: Supplier Quality--what constitutes a reject?

I prefer to handle things as exceptions ....

Hmmmm....the risk involved in that is getting expertise in putting out fires.
The idea is to eliminate the risk of fires and not to be an expert at putting fires out.
 
J

jasonb067

Re: Supplier Quality--what constitutes a reject?

I know this sounds all lofty but the purpose should be to identify waste or non-conformance or anything else costly to business and reduce/elimnate it.

Whatever measure you use should be standard and then valid to contributing to the common goal.

To be more specific for our situation. We measure parts in RI, In process etc. whatever the actual rejected number is counted toward PPM regardless of where it was found. This is all done as part of a larger more complicated score card process I had explained here...

Supplier Certification Programs


The real point is to define a standard form of measurement and apply it and use it for the goal of reducing costs. Having a ppm value just to have it should not be the goal.

Also, there are always going to be situations that require consideration (as described in post 4). It is not going to add any value to improvement to jack up a ppm rating in that situation.

I hope that my opinion helps.
 
R

RosieA

Thanks for everyone's input.

I don't think that this is an issue with measurement for the sake of measurement, but more of an internal disagreement on what should be counted as a reject.

Like Tom, we have issues that get counted against a supplier that really have it's roots within my company. For example, worn casting patterns that we don't want to spend money on replacing then results in castings not being to spec and the supplier being blamed.

It's those kind of issues that I want to get weeded out of the metrics, because the corrective action lies within my company, not at the supplier.
 
D

DrM2u

Here are some questions that hopefully would help a little:
1. Are your plants registered individually or under an umbrella cert?
1.a If they are individually registered, are they in compliance with their established receiving protocols (sample size, frequency, inspection methods, etc)? If they are in compliance then what is the problem?!?
1.b If the plants are under an umbrella cert then is there a common receiving inspection protocol or each plant is given the liberty to establish their own?
1.b.i If there is a common inspection protocol then why is there a discrepancy in the results? Is the protocol implemented uniformly (i.e. same frequency, sample sizes, inspection equipment, qualified personnel, etc)?
1.b.ii If each plant is given its freedom to establish its own protocols then see the questions for 1.a above. Is there any value in having a common receiving protocol?!?

Regarding the definition of a reject, I suggest that you start with the definiton of Non Conforming Product in ISO 9000. After that it is a matter of deciding if to reject the product or accept it as is.
 
J

JRKH

I agree with the earlier posts that suggest seperating product non-conformance from adminstrative (paperwork non-conformance). To me, including paperwork non-conformances not only "inflates" the PPM, it destroys the value of the measure.
If a need is seen to track paperwork issues, then create a seperate measure for this.

I mean - consider this. If a customer "supplies" you with a purchase order for 1000 parts and an error is dicovered in some part of the purchase order does this error count as part of your ppm? If it were, how sould it be counted. 1 defect (the paperwork) or 1000 defects since that's how many parts are involved??

It's a silly example but illustrates how silly some of these things can get.

Peace
James
 
R

RosieA

Here are some questions that hopefully would help a little:
1. Are your plants registered individually or under an umbrella cert?

They are currently individually certified, but moving towards a common cert. (which is why I'm attemptng to consolidate our practices)
1.a If they are individually registered, are they in compliance with their established receiving protocols (sample size, frequency, inspection methods, etc)? If they are in compliance then what is the problem?!?

Yes, they are in compliance. but the plants have varying philosophies regarding receiving inspection.
1.b If the plants are under an umbrella cert then is there a common receiving inspection protocol or each plant is given the liberty to establish their own?
NA at the moment, but needs to be worked out for the future
1.b.i If there is a common inspection protocol then why is there a discrepancy in the results? Is the protocol implemented uniformly (i.e. same frequency, sample sizes, inspection equipment, qualified personnel, etc)?
these plants make different products and have been acquired over the years by the parent company. They were allowed to operate independently and develop their own practices until recently.
1.b.ii If each plant is given its freedom to establish its own protocols then see the questions for 1.a above. Is there any value in having a common receiving protocol?!?
Yes, there's value, and that's what I'm trying to do. but I need to understand if there is a commonly accepted practice for what should be counted in the supplier reject number first.
 
Top Bottom